Changes between Initial Version and Version 1 of Ticket #2312


Ignore:
Timestamp:
Dec 1, 2023, 9:03:38 AM (5 months ago)
Author:
Nicklas Nordborg
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #2312 – Description

    initial v1  
    33It could be useful to have more filtering capabilities. For example, in the list of patients we could then list child derived bioassays with subtype !AlignedSequences and Pipline=DNA/Normal/WGS. The column would then only show us alignments from normal DNA samples and not from the tumor or RNAseq pipelines.
    44
    5 In theory it would be possible to use a very complex filter, but in practice we need to encode everything into the column definition string. This is currently `/DIRECTION/ITEMTYPE/subtype-id/property`. We need to add the extra filter information into this string without breaking existing definitions. Since the subtype-id part is always a numeric ID it would relatively easy to add more information to this field. For example: `/DIRECTION/ITEMTYPE/subtype-id:filter-def/property`
     5In theory it would be possible to use a very complex filter, but in practice we need to encode everything into the column definition string. This is currently `/DIRECTION/ITEMTYPE/subtype-id/property`. We need to add the extra filter information into this string without breaking existing definitions. Since the subtype-id part is always a numeric ID it would relatively easy to add more information to this field. For example: `/DIRECTION/ITEMTYPE/subtype-id:filter-def/property` where we need make sure that `filter-def` is encoded in a way that doesn't break other parsing (eg. it can't contain '/').
    66
    7 where we need make sure that `filter-def` is encoded in a way that doesn't break other parsing (eg. it can't contain '/').
     7To begin with I think that using a single annotation for filtering should be powerful enough.